'LatentDiffusion' object has no attribute 'lora_layer_mapping' after today git pull stable-diffusion-webui

[neskweek]

Hello
after regular git pull I now get, on each generation :

Traceback (most recent call last):
  File "E:\SD\stable-diffusion-webui\extensions\a1111-sd-webui-locon\scripts\..\..\..\extensions-builtin/Lora\lora.py", line 218, in load_loras
    lora = load_lora(name, lora_on_disk.filename)
  File "E:\SD\stable-diffusion-webui\extensions\a1111-sd-webui-locon\scripts\main.py", line 370, in load_lora
    is_sd2 = 'model_transformer_resblocks' in shared.sd_model.lora_layer_mapping
  File "E:\SD\stable-diffusion-webui\venv\lib\site-packages\torch\nn\modules\module.py", line 1614, in __getattr__
    raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'LatentDiffusion' object has no attribute 'lora_layer_mapping' 

running on windows 11 with RTX 3080
COMMANDLINE_ARGS= --xformers --listen

[rangedreign]

I also get this error, running on windows 10 with gtx 1660 ti.
COMMANDLINE_ARGS= --theme dark --autolaunch --precision full --no-half --no-half-vae --medvram --disable-safe-unpickle --reinstall-torch --xformers --always-batch-cond-uncond

[hoodady]

loading Lora Z:\sd\stable-diffusion-webui\models\Lora\locon_angelina_v1_from_v3_64_32.safetensors: AssertionError ?it/s]
Traceback (most recent call last):
File "Z:\sd\stable-diffusion-webui\extensions-builtin\Lora\lora.py", line 222, in load_loras
lora = load_lora(name, lora_on_disk.filename)
File "Z:\sd\stable-diffusion-webui\extensions-builtin\Lora\lora.py", line 192, in load_lora
assert False, f'Bad Lora layer name: {key_diffusers} - must end in lora_up.weight, lora_down.weight or alpha'
AssertionError: Bad Lora layer name: lora_unet_down_blocks_0_downsamplers_0_conv.lora_mid.weight - must end in lora_up.w

I get this random error? whats going on?

[Kadah]

The extension seems to be broken in a few ways with upstream A1111 changes, though I haven't gotten any of them except the following one.

Another breaking change in dev is AUTOMATIC1111/stable-diffusion-webui@39ec4f0, which I've made a RP for: #39

[duhast123]

the previous version
scripts\main.py
Add these lines:

**def assign_lora_names_to_compvis_modules(sd_model):
lora_layer_mapping = {}

for name, module in shared.sd_model.cond_stage_model.wrapped.named_modules():
    lora_name = name.replace(".", "_")
    lora_layer_mapping[lora_name] = module
    module.lora_layer_name = lora_name

for name, module in shared.sd_model.model.named_modules():
    lora_name = name.replace(".", "_")
    lora_layer_mapping[lora_name] = module
    module.lora_layer_name = lora_name

sd_model.lora_layer_mapping = lora_layer_mapping**

def load_lora(name, filename):
print('locon load lora method')
lora = LoraModule(name)
lora.mtime = os.path.getmtime(filename)

sd = sd_models.read_state_dict(filename)

**if not hasattr(shared.sd_model, 'lora_layer_mapping'):
    assign_lora_names_to_compvis_modules(shared.sd_model)**

[NoppaiKohai]

I'm getting this error too, anyone figure out how to fix it

[aetherwu]

the previous version scripts\main.py Add these lines:

**def assign_lora_names_to_compvis_modules(sd_model): lora_layer_mapping = {}

for name, module in shared.sd_model.cond_stage_model.wrapped.named_modules():
    lora_name = name.replace(".", "_")
    lora_layer_mapping[lora_name] = module
    module.lora_layer_name = lora_name

for name, module in shared.sd_model.model.named_modules():
    lora_name = name.replace(".", "_")
    lora_layer_mapping[lora_name] = module
    module.lora_layer_name = lora_name

sd_model.lora_layer_mapping = lora_layer_mapping**

def load_lora(name, filename): print('locon load lora method') lora = LoraModule(name) lora.mtime = os.path.getmtime(filename)

sd = sd_models.read_state_dict(filename)

**if not hasattr(shared.sd_model, 'lora_layer_mapping'):
    assign_lora_names_to_compvis_modules(shared.sd_model)**

It works for me.
Let me clear the code for others:

Add one function before 'def load_lora'

def assign_lora_names_to_compvis_modules(sd_model):
    lora_layer_mapping = {}

    for name, module in shared.sd_model.cond_stage_model.wrapped.named_modules():
        lora_name = name.replace(".", "_")
        lora_layer_mapping[lora_name] = module
        module.lora_layer_name = lora_name

    for name, module in shared.sd_model.model.named_modules():
        lora_name = name.replace(".", "_")
        lora_layer_mapping[lora_name] = module
        module.lora_layer_name = lora_name

    sd_model.lora_layer_mapping = lora_layer_mapping

Replace first a few lins o 'def load_lora'

as

def load_lora(name, lora_on_disk):
    print('locon load lora method')
    lora = LoraModule(name, lora_on_disk)
    lora.mtime = os.path.getmtime(lora_on_disk.filename)
    sd = sd_models.read_state_dict(lora_on_disk.filename)

    if not hasattr(shared.sd_model, 'lora_layer_mapping'):
        assign_lora_names_to_compvis_modules(shared.sd_model)

    is_sd2 = True  // I don't really know what this is for anther function ask for it.

The 'lora_on_disk' part is compatible with latest PR:
https://github.com/KohakuBlueleaf/a1111-sd-webui-locon/pull/39/files

The error reported disppeared.

AttributeError: 'LatentDiffusion' object has no attribute 'lora_layer_mapping'

However one new errror emerged:

Failed to match keys when loading Lora models\Lora.....

Magically, it does not effect final result. All LoRA models works fine to me...